Conductor Short Circuit Protection (10-6-2K)
 

 

Download this article [ Word™ ] [ PDF ]

Conductor Sizing and Overcurrent Protection Continued….

As a follow-up to Mike Holt's article on conductor sizing and overcurrent protection (http://www.mikeholt.com/Newsletters/conductorsize-protection.htm), this article will address an additional criteria for conductor sizing that is often overlooked, short-circuit protection of conductors.

Short-circuit protection of conductors depends upon the available fault current and the type and speed of the overcurrent protection device.  If one looks at the fine print note in Section 240-1, it alludes to the fact that other requirements are present for conductor protection as well and reads:

"FPN: Overcurrent Protection for conductors and equipment is provided to open the circuit if the current reaches a value that will cause an excessive or dangerous temperature in conductors or conductor insulation.  See also Section 110-9 for requirements for interrupting rating and Section 110-10 for requirements for protection against fault currents."

Section 110-9 addresses the interrupting rating of the overcurrent device required based upon short-circuit current.  Section 110-10 addresses the overcurrent protective device, component short-circuit current ratings and protection of the circuit components by the overcurrrent protective device to prevent "excessive damage".  This is especially important because it is not just protection against overloads, it is also protection from short-circuits. 

Introduction:

               Now the question remains, how can we verify protection of conductors from "extensive damage" against short-circuits? The answer depends upon the interpretation of "extensive damage". Information on conductor short-circuit current is available based upon the amount of damage the user allows.  One level of protection is based upon the insulation of the conductor.  Formulas and charts from the IEC and the Insulated Cable Engineers Association (ICEA) are available to illustrate this level of protection.  The charts determine how much time and current is required to raise the temperature of the conductor from the operating temperature to a temperature where slight damage to thermoplastic insulation occurs.  For conductors with 75 deg. C thermoplastic insulation, the damaging temperature is typically 150 deg. C.  Another level of protection is based on the Soares "validity" rating.  The validity rating corresponds to the amount of current required to cause the copper to become loose under a lug after the conductor has had a chance to cool back down.  This validity rating is based upon raising the copper temperature from 75 deg. C to 250 deg. C, which is the annealing temperature of copper.  In addition, a third level of protection promoted by Onderdonk allows the calculation of the current necessary to cause the conductor to melt (75 deg. C to 1,083 deg. C). 

Table 1 illustrates the conductor rating for 5 seconds as well as the maximum I2t (ampere2 seconds) rating based upon conductor size and the conductor damage level (ICEA, Soares or Onderdonk). 

However, depending upon the damage level selected, the overcurrent protective device selected and the resulting opening time, the amount of current the conductor can handle will need to be adjusted.  If we select the ICEA insulation damage level as the desired protection level for conductors, we can establish more usable information with respect to the clearing time of the overcurrent protective device.  Since some devices such as airframe/power circuit breakers exist which can have short-time delays up to 30 cycles, we must analyze the opening time up to this level.  Table 2 shows the amount of current the conductor can handle based upon different clearing times (up to 30 cycles) with respect to the ICEA insulation damage level.

Table 1:  Comparison of Conductor Current Ratings (Based on RMS Amperes)

   

5 Second Rating (Amps)

I2t Rating (Amperes Squared Seconds)

Cond

Cond

ICEA

Soares

Onderdonk

ICEA

Soares

Onderdonk

Size

Area

Insulation

Annealing

Melting

Insulation

Annealing

Melting

Circ.  Mils

150 Deg. C

250 Deg. C

1083 Deg. C

150 Deg. C

250 Deg. C

1083 Deg. C

14

4110

97.31

139.84

251.35

47346

97774

315881

12

6530

154.61

222.18

399.34

119517

246812

797382

10

10380

245.76

353.17

634.79

301994

623640

2014813

8

16510

390.90

561.74

1009.68

764007

1577732

5097229

6

26240

621.27

892.79

1604.72

1929882

3985352

12875605

4

41740

988.25

1420.16

2552.63

4883238

10084258

32579535

3

52620

1245.85

1790.34

3218.00

7760768

16026576

51777572

2

66360

1571.17

2257.83

4058.27

12342860

25488942

82347942

1

83690

1981.48

2847.46

5118.10

19631350

40540229

130974616

1/0

105600

2500.23

3592.93

6458.01

31255818

64545637

208529659

2/0

133100

3151.33

4528.59

8139.79

49654561

102540438

331280679

3/0

167800

3972.91

5709.22

10261.88

78919977

162975743

526530956

4/0

211600

5009.94

7199.46

12940.49

125497294

259161440

837281168

250

250000

5919.11

8505.98

15288.86

175179408

361758775

1168745667

300

300000

7102.93

10207.18

18346.63

252258347

520932636

1682993761

350

350000

8286.76

11908.38

21404.40

343351639

709047199

2290741508

400

400000

9470.58

13609.57

24462.17

448459284

926102464

2991988908

500

500000

11838.22

17011.97

30577.71

700717631

1447035100

4674982669


Table 2: Maximum Short-Circuit Current Rating In Amperes (Per ICEA Insulation Damage)
Maximum Short-Circuit Current Rating In RMS Amperes
Cond Cond 1/2* 1 2 3 6 12 18 24 30
Size Area Cycles Cycle Cycle Cycle Cycle Cycle Cycle Cycle Cycle
Circ. Mils 0.0083 0.0167 0.0333 0.0500 0.1 0.2 0.3 0.4 0.5
  Seconds Seconds Seconds Seconds Seconds Seconds Seconds Seconds Seconds
14 4110 2384 1685 1192 973 688 487 397 344 308
12 6530 3787 2678 1894 1546 1093 773 631 547 489
10 10380 6020 4257 3010 2458 1738 1229 1003 869 777
8 16510 9575 6771 4788 3909 2764 1954 1596 1382 1236
6 26240 15218 10761 7609 6213 4393 3106 2536 2197 1965
4 41740 24207 17117 12104 9883 6988 4941 4035 3494 3125
3 52620 30517 21579 15259 12459 8810 6229 5086 4405 3940
2 66360 38486 27213 19243 15712 11110 7856 6414 5555 4968
1 83690 48536 34320 24268 19815 14011 9907 8089 7006 6266
1/0 105600 61243 43305 30621 25002 17679 12501 10207 8840 7906
2/0 133100 77192 54583 38596 31513 22283 15757 12865 11142 9965
3/0 167800 97316 68813 48658 39729 28093 19865 16219 14046 12563
4/0 211600 122718 86775 61359 50099 35426 25050 20453 17713 15843
250 250000 144988 102522 72494 59191 41854 29596 24165 20927 18718
300 300000 173986 123026 86993 71029 50225 35515 28998 25113 22461
350 350000 202983 143531 101492 82868 58596 41434 33831 29298 26205
400 400000 231981 164035 115990 94706 66967 47353 38663 33484 29949
500 500000 289976 205044 144988 118382 83709 59191 48329 41854 37436
* When comparing these values for 1/2 cycle with the available RMS symmetrical fault current, multiply the available RMS symmetrical by 1.3 to account for asymmetry during the first half cycle.

The next step to analyzing protection of conductors against short circuits depends upon the overcurrent device selected and available fault current.  The overcurrent protective device can either be a current limiting fuse or circuit breaker (current liming or non-current limiting).

Current Limiting Devices:

A device, which opens in one-half cycle or less, may or may not be classified as "current-limiting".  The device can be labeled current limiting only if it meets the requirements of UL/CSA (489/22.2 No. 5 for molded case circuit breakers or UL/CSA 248 for fuses).

The requirement for molded case circuit breakers is that the device must limit the asymmetrical fault current to a value below the equivalent symmetrical fault current.  If the circuit breaker is not current-limiting, but clears within approximately one-half cycle, the available symmetrical fault current must be multiplied by a factor of 1.3 to account for asymmetry.  Even if the molded case circuit breaker is current-limiting, the degree of current limitation is typically less than a rejection type fusible device.

The requirements for current-limiting fuses are based on the fuse type (Class) and upon the maximum I2t let-through as shown in Table 3.  If the device meets the performance criteria, it meets the standard for that class of fuse and can be listed/certified per UL/CSA.

The values shown in the table to the right can be used to compare the maximum I2t let-through for the fuse to the I2t rating of the conductor.  As long as the I2t let-through of the fuse is less than the I2t rating of the conductor, the conductor is protected under short-circuit conditions. 

Table 3:

Max Let-Through (Ampere2 Seconds) per UL 248 - Table A

Fuse

I2t Max Let-through Value

Class

50

100

200

kA

kA

kA

 Class J

30

7,000

7,000

7,000

60

30,000

30,000

30,000

100

60,000

80,000

80,000

200

200,000

300,000

300,000

400

1,000,000

1,100,100

1,100,100

600

2,500,000

2,500,000

2,500,000

 Class RK1

30

10,000

10,000

11,000

60

40,000

40,000

50,000

100

100,000

100,000

100,000

200

400,000

400,000

400,000

400

1,200,000

1,200,000

1,600,000

600

3,000,000

3,000,000

4,000,000

 Class RK5

30

50,000

50,000

50,000

60

200,000

200,000

200,000

100

500,000

500,000

500,000

200

1,600,000

1,600,000

2,000,000

400

5,200,000

5,000,000

6,000,000

600

10,000,000

10,000,000

12,000,000

 Class T

30      300V

3,500

3,500

3,500

600V

7,000

7,000

7,000

60      300V

15,000

15,000

15,000

600V

30,000

30,000

30,000

100    300V

40,000

40,000

40,000

600V

60,000

80,000

80,000

200    300V

150,000

150,000

150,000

600V

200,000

300,000

300,000

400    300V

550,000

550,000

550,000

600V

1,000,000

1,100,000

1,100,000

600    300V

1,000,000

1,000,000

1,000,000

600V

2,500,000

2,500,000

2,500,000

An example examining both the phase and grounding conductor is given on the following page.

Short-Circuit Protection of Wire and Cable:

Fusible Systems

The circuit shown originates at a distribution panel where 40,000 amperes RMS symmetrical are available. The # 10 THW copper conductor is protected by a Bussmann LOW-PEAK® fuse sized per NEC 240-3 (30A maximum for # 10 conductor). 

Table 1, shows the I2t withstand rating of # 10 THW copper to be 301,994 000 ampere2 seconds per the ICEA Insulation damage level.  Since the Bussmann LOW-PEAK® is a Class RK1 current limiting fuse, the maximum I2t let-through per UL/CSA 248 can be found per the table on the previous page.  The maximum I2t let-through for a Class RK1 fuse at a fault not greater than 50,000A is 10,000 ampere2 seconds.  Since the I2t let-through of the fuse, 10,000, is considerably less than the I2t withstand of the # 10 conductor, 301,994, the conductor is protected against short-circuits.  In addition, since the minimum grounding conductor per NEC 250-122 is also # 10, the grounding conductor is protected as well.

Table 4, on the next page, compares the maximum I2t let-through for the fuse with the withstand of the conductor.  If the I2t withstand of the conductor is greater than the I2t let-through of the fuse; the conductor will be protected.  The table shows the minimum size conductor able to be protected under short-circuit conditions by the fusible device. 

Circuit Breaker Systems

In the previous example a 30A, Class RK1 fuse was protecting a # 10 conductor.  If the 30A fusible device were replaced with a 30A, molded case circuit breaker with a clearing of 1/2 cycle would the # 10 conductor be protected? 

Since the I2t let-through of the circuit breaker is not known, Tables 2 must be used.  If the 30A circuit breaker is current limiting, at a 40,000A fault, with a clearing time of 1/2 cycle, per UL 489, the let-through current could be as high as 40,000A RMS and still be marked current limiting.  If the device was not current limiting, the let-through current could be as high as 40,000 X 1.3, or 52,000A.  The maximum short-circuit current rating of # 10 conductor is 6,020A for 1/2 cycle (per ICEA).  Since the let-through current of either the current limiting or non-current limiting circuit breaker could be much greater than the short-circuit current rating of the conductor, protection can not be assured for either the phase or equipment grounding conductor.

Table 5, on the next page, can be utilized to analyze conductor protection by current-limiting and non-current limiting circuit breakers.  Note in the tables, the size of the conductor has increased due to the limited degree of current-limitation or the increased opening time of the device (compared to a fusible system).  When using molded case circuit breakers, insulated case circuit breakers or low voltage power circuit breakers, conductors must be carefully analyzed for protection.  This is especially true for equipment grounding conductors since the NEC allows for reduced sizing of the equipment grounding conductor despite the fact that available ground fault currents can be equal to or perhaps greater than the available three phase short-circuit currents. 

Table 4:

Conductor Short-Circuit Protection (ICEA Values)

Conductor Short-Circuit Protection (ICEA Values)

Minimum Conductor Size, Based on UL 248

Minimum Conductor Size, Based on UL 248

Fuse

Short-Circuit Current (RMS Amperes)

Fuse

Short-Circuit Current (RMS Amperes)

Class

50

100

200

Class

50

100

200

kA

kA

kA

kA

kA

kA

 Class RK1

 Class J

30

14

14

14

30

14

14

14

60

14

12

12

60

14

14

14

100

12

12

12

100

14

14

14

200

8

8

8

200

12

10

10

400

6

6

6

400

8

8

8

600

4

4

4

600

6

6

6

 Class RK5

 Class T

30

12

12

12

30       300V

14

14

14

60

10

10

10

600V

14

14

14

100

8

8

8

60       300V

14

14

14

200

6

6

4

600V

14

14

14

400

3

3

3

100     300V

14

14

14

600

2

2

2

600V

14

14

14

200     300V

12

12

12

600V

12

10

10

400     300V

10

10

10

600V

8

8

8

600     300V

8

8

8

600V

6

6

6

Table 5:

Circuit

Short-circuit Current (RMS Amperes)

Breaker's

(Clearing

10

25

50

100

200

or STD***)

kA

kA

kA

kA

kA

Setting

1/2 cycle C.L.*

#6

#3

1/0

4/0

350

1/2 cycle N.C.L.**

#6

#2

2/0

250

500

 1 Cycle

#6

#2

2/0

250

500

 3 Cycle

#3

1/0

4/0

500

900

 6 Cycle

#2

3/0

300

600

-

 12 Cycle

1/0

4/0

500

900

-

 18 Cycle

1/0

300

600

-

-

 24 Cycle

2/0

300

600

-

-

 30 Cycle

3/0

350

700

-

-

*  C.L. = Current Limiting Circuit Breaker.  Values are based upon UL 489/CSA 22.2 No. 5

requirements to limit short-circuit currents to the symmetrical values.

** N.C.L. = Non-Current Limiting Circuit Breaker

*** STD = Short Time Delay

The above material was supplied by: Daniel R. Neeser, Manager, Technical Sales Cooper Bussmann

E-mail:  dneeser@buss.com, Website:  http://www.bussmann.com


Circuit Breaker Manufacture Response

As I suspected, this article from Bussmann is an attempt to "scare" people from using circuit breakers.  (Bussmann has numerous articles and videos that are produced using SELECTIVE DATA with the sole intention of discrediting the use of breakers.)  The fact is, that no matter how many theoretical calculations are performed, Power Circuit Breakers and Molded-Case Circuit breakers are lab-tested to UL standards WITH THE POWER CONDUCTORS AND LUGS attached to the circuit.  You can refer to the following standards:
UL1558 (standard for metal enclosed circuit breaker switchgear)
UL1066 (standard for power circuit breakers)
UL891 (standard for dead-front switchboards)
UL489 (standard for molded-case circuit breakers)

Note: Bussmann comments are contained in blue: While the above standards do test performance of the overcurrent device & assembly with the conductor as an integral part of the test, it may not truly prove proper protection.  While UL 489, does not allow ejection of wire from the wire connector, burn off of leads or damage to conductor insulation, this is only for two short-circuit tests.  What can happen over the life of the cable?  How many short-circuits will the circuit breaker protect the conductor, is it two or more?  UL 489 does not articulate how insulation damage is evaluated.  Slight damage to insulation may not be easily seen by visual inspection.

In addition, UL 489 does not require the circuit breaker be tested with the size of equipment grounding conductor that it may be called upon to protect.  For example, a 60 ampere molded case circuit breaker is not tested with a #10 equipment grounding conductor. 

Additionally, in the bus shot tests for circuit breakers 100A or less, where the 4'10" of rated conductor is not used (#1 is used instead), the condition of the wire after the test is not even part of the pass/fail criteria for the circuit breaker.   As a side note, it is interesting that in the UL 489 bus shot test, the circuit breaker does not have to "relatch, reclose, or otherwise reestablish continuity".


The reason some circuit breakers are intentionally designed with a short-time rating (typically large power breakers, above 800A), is to allow the downstream (usually current-limiting) feeders to clear the fault, so the main breaker stays closed to power the rest of the facility (Power System Coordination).  This is very critical in situations where losing power will cause loss of life, or where a critical process will be shut down unexpectedly (petro-chemical, steel mills, data centers, etc.) 

Bussmann comment to the above: Like the writer indicated, short-time delay is required in circuit breakers because the upstream or line-side circuit breakers need to hold in while the downstream circuit breakers open the fault.  However, the delay is required because the downstream breakers are NOT usually current-limiting (by definition would have to clear within 1/2 cycle).  If they were current-limiting, there would probably be no need for short-time delay.  As pointed out in the article, the concern is encountered when the fault happens between the downstream feeder circuit breaker and upstream circuit breaker.  Then the breaker intentionally holds in on the fault and additional damage can occur.

This article also seems to overlook the fact that the wire itself tends to limit the fault current, due to conductor impedance.  Referring to the "theoretical" example illustrated in the article, if a fault occurs 10 feet away from a source that can deliver 40kA using #10 wire, the cable alone will limit the fault to about 10kA.  The author seems to think that 40kA gets passed through the conductor with zero impedance.  (For that matter, if the fault occurs in the wire, the wire will need to be replaced anyway...)

Bussmann comment to the above: The writer is absolutely correct; the wire will limit the fault.  If one would prefer to use the fault current at the load in lieu of the value at the beginning of the run that is fine, but not all faults will occur at the end of a run of cable.  Faults can occur anywhere along the run of cable, and while that cable will need to be replaced, it would not be desirable to damage any of the other cables in that raceway.  Either way, my tables, such as table 5, will still be valid, except the available fault current would reflect the fault at the load.  The important thing to remember is we are trying to prove protection, based upon physics.  The challenging part with a molded case (non-current limiting) circuit breaker is that often the data does not exist or is not available to "prove" protection.  Because of this, often, protection is assumed based upon testing such as UL 489.  However, these tests attempt to approximate "real-world", they do not necessary assure "real-world" results as stated above.  The simple solution would be if UL 489 had test criteria and limits for energy let-through (I2t & Ip) for circuit breakers such as UL 248 for various fuse classes, but UL 489 does not.

Other factors to consider: Line-connected fuses DO NOT offer ground-fault protection. 

This was not made clear in the "Fusible Systems" example.  In the unlikely event that it is a "bolted" ground fault, and the prospective short-circuit current matches that of the primary circuit, only then will the fuse blow.  This, incidentally, will blow only one fuse in a 3-phase circuit, allowing the system to "single-phase", causing damage to downstream loads, such as motors.  In the practical application, bolted faults are the exception, not the norm (although the system MUST be designed to account for this situation).  Arcing faults in motors are the more common fault condition, which are normally low-magnitude.  More sensitive ground-fault equipment is required for this condition.  Circuit breakers offer this protection, along with the ability to open all 3 phases on a single line-ground fault.

Bussmann comment to the above: Neither fused switches nor circuit breakers will protect against all low-level arcing faults.  However, both fused switches and circuit breakers can be equipped with ground fault protection.  For fused switches it requires a shunt trip along with the ground fault sensing equipment.  Circuit breakers are easily equipped with integral ground fault protection.  However, to think that all circuit breakers will have ground fault protection just because it can be integral to the circuit breaker is unrealistic.  In fact, typically only the main device will have ground fault protection, every other device in the distribution system normally does not.  Thus, it may be a mute point. 

In addition, if protection against single-phasing for motors is required, the proper protection is not from a fused switch or circuit breaker, it is a properly sized overload relay or an overload relay with single-phase protection.


Where does Bussmann get the data for Table 5. Is this from UL, the NEC, IEEE, or ICEA?  It appears this data is fabricated using zero lead-length conductors, with zero cable impedance.

Bussmann comment to the above: Table 5 is comparing the possible current let-through based upon the opening time by a molded case circuit breaker in relation to the conductor and selecting the smallest that will be able to be protected.  This assumes a zero lead length, for simplicity sake.  However, if the available fault current was assumed as the value at the load, this could be used instead and would take into account the lead length. 

Until real-world data is used, I recommend that you remove this newsletter from your outstanding web site.  I very much enjoy the letters and examples that you supply to the industry.

Bussmann comment to the above: As mentioned before, this is as close to "real-world" as possible, based upon physics formulas.  Bussmann has videotapes of various types of overcurrent devices protecting or attempting to protect conductors under short circuit conditions, which validate, at least partially, the material in my article.  Bussmann District Sales Engineers are more than happy to show these video taped tests upon request.


If you have any additional questions, please feel free to contact me!!

Wayne Celeste
Cutler-Hammer District Application Engineer - Edison, NJ
celesw@ch.etn.com

Mike Holt’s Comment:

Nothing every changes, circuit breaker and fuse guys are always battling it out and this is good for the industry, it keeps every body on their toes.

  Go to top of page
Newsletter Registration   |   Stay Connected:
 

888.NEC.CODE (632.2633) 3604 PARKWAY BLVD, STE 3, LEESBURG FL 34748  

Tell a Friend About This Site

  NEC® and National Electrical Code® are registered trade marks
of the National Fire Protection Association (NFPA).
  ©Copyright 2011 Mike Holt Enterprises, Inc.